OSF HealthCare has long delivered expert neurovascular care across its 18-hospital system, headquartered in Peoria, Illinois.
But as stroke volumes increased and care complexity grew, so did the pressure to make faster, more precise decisions – especially when managing transfers across multiple sites.
THE CHALLENGE
With limited ICU capacity at its comprehensive stroke centers, the health system needed to ensure only the patients who would truly benefit from escalation were transferred. At the same time, variations in imaging workflows and system-wide staffing constraints made it harder to consistently achieve the speed and coordination these cases demand.
"We know care variation is closely tied to poorer outcomes and higher costs – challenges that are especially true in stroke care, where every delay increases the threat of mortality," said Dr. Deepak Nair, vice president of the neuroscience service line at OSF HealthCare.
"Beyond imaging itself, we recognized the need for a more streamlined and connected communication process – one that could bring together all stakeholders, across hospitals and disciplines, in real time.
"In a complex, fast-moving case, the ability to align teams quickly is just as critical as the scan itself," he continued. "We also faced persistent challenges around quality control and workforce capacity, driven by a nationwide shortage of radiologists. These gaps made it even more important to augment our teams with tools that could support consistency, accuracy and collaboration at scale."
PROPOSAL
OSF HealthCare studied technology from RapidAI, a vendor of artificial intelligence-driven medical imaging analysis and coordinated care systems. The vendor proposed a platform built to tackle both the clinical and operational challenges OSF was facing.
"At its core, the platform promised to accelerate door-to-decision and door-to-treatment times by delivering fast, consistent imaging insights to help guide critical transfer and treatment decisions," Nair explained.
"It was designed to support clinicians of varying experience levels in making more confident calls – whether they were identifying vessel occlusions, evaluating hemorrhages or spotting subtle findings that may otherwise be missed."
The goal was to reduce care variation while increasing the precision of every decision.
"The platform also addressed our need for consistency during unpredictable conditions: high volumes, off-hours or limited radiology staffing," Nair said. "By providing AI-driven support 24/7, the vendor promised to bring a baseline of quality and reliability to every case, regardless of when or where it originated.
"This was essential not only for improving patient care, but also for helping us operate more efficiently as a system – preserving ICU beds, avoiding unnecessary transfers and focusing attention where it was truly needed," he added.
Finally, the long-term vision was attractive to OSF. The vendor's roadmap aligned with that of the health system – offering expansion into additional use cases like aneurysm management, hemorrhagic stroke and 3D reconstruction.
"The platform wasn't just a short-term fix – it was a scalable investment in how we deliver coordinated, high-performing care now and into the future, even beyond neurovascular care," he noted.
MEETING THE CHALLENGE
OSF Healthcare deployed the RapidAI system across its neuroscience network to help address care variation, streamline triage and support timely treatment decisions.
Emergency physicians, neurologists, neurointerventionalists and transfer coordinators all use the platform to evaluate imaging and determine the most appropriate next step – whether it be rapid escalation, local management or additional surveillance.
The system-wide use of various modules of the vendor's technology gave OSF teams fast, objective insights that gave spoke-and-hub sites a leveled playing field, Nair said.
For example, Rapid Aspects played a key role in reducing interpretation variability. By providing standardized scoring within seconds, it helped elevate consistency across all readers – bringing general radiologists and ED providers closer to the level of subspecialty neuro-interpretation, Nair explained. This was particularly impactful in settings where on-call neuro expertise wasn't immediately available.
"Because RapidAI integrated directly into existing systems like our Epic EHR and our imaging infrastructure, adoption was seamless," he noted. "Teams didn't have to learn an entirely new workflow, they simply got faster, clearer data within the infrastructure they were already used to. That ease of access was essential in driving widespread use and showing value quickly."
RESULTS
Since implementing the medical imaging analysis and coordinated care technology, OSF Healthcare has seen a 17% increase in intracranial hemorrhage (ICH) patients identified per month, along with a 34% increase in ICH procedure volume from 2023 to 2024.
The vendor's ICH module has enabled OSF clinicians to quickly detect suspected bleeds, while another module quantifies blood volume with precision, Nair explained.
"This level of insight supports earlier decision making around whether and when to treat or transfer, helping us act with greater confidence," he said.
On the ischemic stroke side, the improvements have been even more striking. Despite a 9% drop in emergency department visits, OSF saw a 155% increase in stroke diagnoses and a 347% increase in mechanical thrombectomy procedures.
While many factors contribute to these trends, the vendor's AI-powered tools for non-contrast CT, CTA and perfusion) have been central to this progress, Nair reported.
"These modules provide timely, detailed imaging insights and enable better coordination across teams – so decisions are not only faster, but also smarter," he said. "Time to treatment remains critical, but what's equally important is making the right decision for each patient. AI helps our clinicians see more in every scan, which leads to more targeted, effective care.
"This increase in mechanical thrombectomies means 111 more patients received timely, targeted care – translating to an estimated $2.6 million in additional annual revenue," he reported. "When you factor this into the broader evaluation of a clinical AI platform's ROI, the value becomes clear: It's not just about improving care quality, it's also about strengthening the sustainability of the service line."
The artificial intelligence technology has helped OSF deliver better outcomes while unlocking the operational and financial performance needed to scale that impact across the system, he added.
ADVICE FOR OTHERS
When evaluating AI-powered radiology, focus first on whether it can truly enhance clinical decision making, Nair advised.
"The value shouldn't just be in flagging abnormalities or accelerating workflows – it should be in giving care teams the information they need to make smarter, more confident decisions for each individual patient," he explained.
"Especially in complex or time-sensitive scenarios, AI should act as a decision support partner, helping reduce uncertainty and variation across all levels of expertise.
"Equally important is selecting a platform, not a one-off tool," he continued. "You need something that integrates seamlessly into existing workflows and systems, whether that's your EHR, imaging infrastructure or communication tools. The right platform should feel like an extension of what your teams already do, not a disruption."
And as clinical needs evolve, whether it's expanding stroke programs, scaling to trauma or adding advanced imaging capabilities, the AI platform should be able to grow with the health system without forcing staff to reinvent the wheel, he added.
"Simplicity, interoperability and scalability are what turn AI from a pilot project into lasting system transformation," he concluded.
Follow Bill's health IT coverage on LinkedIn: Bill Siwicki
Email him: bsiwicki@himss.org
Healthcare IT News is a HIMSS Media publication.
WATCH NOW: Helpful tips on smart rooms from OhioHealth
